Schizophrenia is a relatively frequent and disabling disease of variable expression and unknown, but perhaps heterogeneous cause. Definition of subtypes and pathogenesis-symptom complex relationships are prerequisites to advances in therapeutics and prevention. The objective of our research is to test the hypothesis of heterogeneity by searching for genetically validated subtypes. To this end, we have collected from an epidemiologic sample, a large number of families with more than on affected member. We now propose a genetic study of these families using recombinant DNA methods. In addition to candidate gene and chromosome specific probes, we will use cytogenetic and biochemical polymorphisms to look for linkages and associations which will suggest areas of specific chromosomes to examine with relevant probes. Associations of clinical expression with specific genes would form a useful basis for a classification of the types of schizophrenia.